CVE-2026-90183

Name of the Vulnerable Software and Affected Versions Linux kernel (affected versions not specified)

Description An issue exists in the blk-iolatency component where the io.latency controller can throttle a group that lacks its own latency target. This occurs when a sibling misses its target, causing check scale change() to scale down peers. A peer reaching queue depth one triggers blkcg use delay() during subsequent scale-downs, even if min lat nsec is 0. Because iolatency pd offline() only clears the delay during a nonzero to zero transition, the delay remains set when policy data is freed. This leaves blkg->use delay active and blkcg->congestion count elevated, causing blk cgroup congested() to return true for all tasks in that cgroup and its descendants. Consequently, page cache sync ra() reduces readahead to a single page, page cache async ra() skips readahead entirely, and folio throttle swaprate() schedules a throttle on anonymous folio allocation.

Recommendations At the moment, there is no information about a newer version that contains a fix for this vulnerability.